In 2015, the United Nation General Assembly adopted the 2030 Agenda for Sustainable Development and its 17 Sustainable Development Goals aiming at ending all forms of poverty, fighting inequalities, and tackling climate change. We collected Twitter data about the 2030 Agenda from May 9th to November 9th, 2018. The aim of this work is to obtain a classification of each tweet in the corpus according to the “Information”—“Action” categories, in order to detect whether a tweet refers to an event or it has only an informative-disclosure purpose. It seems particularly interesting to understand how and to what extent people and organizations are playing a more active role in shaping the process of responding locally and internationally to climate change. Explicit intention to act or inform had been captured by hand coding of a randomly selected sample of tweets and then the classification had been extended to the whole corpus through a supervised machine learning method. Overall, our classification supervised model has produced satisfactory results.

Supervised Machine Learning Methods to Disclose Action and Information in “U.N. 2030 Agenda” Social Media Data

Sciandra A.;Surian A.;Finos L.
2021

Abstract

In 2015, the United Nation General Assembly adopted the 2030 Agenda for Sustainable Development and its 17 Sustainable Development Goals aiming at ending all forms of poverty, fighting inequalities, and tackling climate change. We collected Twitter data about the 2030 Agenda from May 9th to November 9th, 2018. The aim of this work is to obtain a classification of each tweet in the corpus according to the “Information”—“Action” categories, in order to detect whether a tweet refers to an event or it has only an informative-disclosure purpose. It seems particularly interesting to understand how and to what extent people and organizations are playing a more active role in shaping the process of responding locally and internationally to climate change. Explicit intention to act or inform had been captured by hand coding of a randomly selected sample of tweets and then the classification had been extended to the whole corpus through a supervised machine learning method. Overall, our classification supervised model has produced satisfactory results.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11577/3388563
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 5
social impact